Searching \ for '[PIC] interlocks needed?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'interlocks needed?'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] interlocks needed?'
2008\10\15@203718 by Dr Skip

picon face
While I await my brand new shiny pickit2, I'm going over a design I have that's
a one-off project. I thought I'd tap the experience of the list on something.

One of the lines has to control a pair of lines for forward/reverse control of
a high torque, high speed motor (simplified version). Major damage would result
if both lines came on at the same time, or if it switched too fast (while in
motion). To prevent simultaneous 'on', it'll drive a SPDT relay - only one
state at a time. Power (through another relay) will only be applied after the
direction relay has been set (timed, not through feedback). Both power and
direction will be driven from a pic. Reversing while in motion will cause much
damage, but it isn't human injury. Direction and on/off are controlled by the pic.

If it were a human injury problem, I'd have to put in some sort of interlock to
prevent problems. The question is, given power glitches, pic issues, power up,
etc, what transients do I need to watch for on the pic outputs or what
experiences tell you that in this case, a secondary mechanism is needed? If the
failure mode of the outputs is to all turn off, no worries then. I'm more
concerned about a situation where power is on and it's running and some
condition causes the other (direction control line) output to change state,
even for a few mS. I'd like to foresee as many contingencies as possible.

I also need to know how outputs respond if power to the pic is lost. Are there
any pulses or transients that outputs have or do they just fall to 0 (in real
experience)?

Thanks,
Skip

2008\10\15@211353 by Xiaofan Chen

face picon face
On Thu, Oct 16, 2008 at 8:37 AM, Dr Skip <spam_OUTdrskipTakeThisOuTspamgmail.com> wrote:
> If it were a human injury problem, I'd have to put in some sort of interlock to
> prevent problems. The question is, given power glitches, pic issues, power up,
> etc, what transients do I need to watch for on the pic outputs or what
> experiences tell you that in this case, a secondary mechanism is needed?

You may have to look at much more than what you described.
This is so called a machine safety application and you may
have to go through certification like SIL 3. Not an easy task
for an small organization.

One link for you for the safety product.
http://www.ab.com/safety/
Disclaim: I am working for this company.

Xiaofan

2008\10\15@213334 by Jinx

face picon face
One fail-safe might be to use the PIC to drive one-shots with regular
pulses. Period between the pulses is shorter than the time-out period
of the one-shot. If the PIC fails, so do the pulses and the one-shot will
time out. There *shouldn't* be power-up glitches if PWRTE,  /mclr,
etc are used properly

You might want to measure the activity of the motor before switching
it. eg detect rpm or use some sort of sense element that you can measure
the voltage of or across

2008\10\15@213532 by Michael Algernon

flavicon
face
>
> On Oct 15, 2008, at 6:37 PM, Dr Skip wrote:
>
> While I await my brand new shiny pickit2, I'm going over a design I  
> have that's
> a one-off project. I thought I'd tap the experience of the list on  
> something.
>
> One of the lines has to control a pair of lines for forward/reverse  
> control of
> a high torque, high speed motor (simplified version). Major damage  
> would result
> --snip--
>
> Thanks,
> Skip
-- If this is one off and major damage could occur, I recommend a  
front end that is not associated with the PIC that prevents disallowed  
states for the motor.
for example:  You could have a  circuit that controls a relay or FET  
that disables power when a reversal command occurs.  It would disable  
power until the motor comes to a halt.  ( I am ignoring any braking  
circuitry you might have ).  If you are interested I could sketch up a  
non-microcontroller design that does this.
If your PIC design has a x% chance of failure and this front end  
circuit has a .01% chance of failure , your total failure rate would  
be approximately .0001x.
Michael

2008\10\16@010032 by Dr Skip

picon face
Thanks for the offer (and all for the replies). I can whip something up, since
it sounds like it's a generally good idea. I felt I should probably do
something like that, but wanted to verify that it just wasn't my mistrust of
the little chip. I've been doing hardware since the time when paper tape
readers were a 'fancy' peripheral (long before the PC) and CPUs were bit-sliced
ALUs in TTL, so I'm glad it wasn't just my Neanderthal, pre-pic bred worries... ;)

-Skip



Michael Algernon wrote:
{Quote hidden}

2008\10\16@100235 by Eoin Ross

flavicon
face
I recently made a 'system' where I had two pumps that could not run at the same time. It ended up with a mechanical interlock between the two contactors, and a delay-off timer that cut power to both start buttons.

In your case I'd still put an interlock in to prevent damage - downtime can cost more than the physical damage in some processes. Depending on what you are driving it might be cheap enough to change out the motor controls to provide the mechanical end.

12 Amp Contactors ~$50 each (With overloads added)
http://web1.automationdirect.com/adc/Shopping/Catalog/Motor_Controls/Fuji_Contactors_-z-_Overloads/9_to_25_Amp/SC-E04-24VAC

Mechanical interlock ~$11
http://web1.automationdirect.com/adc/Shopping/Catalog/Motor_Controls/Fuji_Contactors_-z-_Overloads/9_to_25_Amp/SZ-RM


>>> Dr Skip <.....drskipKILLspamspam@spam@gmail.com> 15 Oct 08 20:37:11 >>>
While I await my brand new shiny pickit2, I'm going over a design I have that's
a one-off project. I thought I'd tap the experience of the list on something.

One of the lines has to control a pair of lines for forward/reverse control of
a high torque, high speed motor (simplified version). Major damage would result
if both lines came on at the same time, or if it switched too fast (while in
motion). To prevent simultaneous 'on', it'll drive a SPDT relay - only one
state at a time. Power (through another relay) will only be applied after the
direction relay has been set (timed, not through feedback). Both power and
direction will be driven from a pic. Reversing while in motion will cause much
damage, but it isn't human injury. Direction and on/off are controlled by the pic.

If it were a human injury problem, I'd have to put in some sort of interlock to
prevent problems. The question is, given power glitches, pic issues, power up,
etc, what transients do I need to watch for on the pic outputs or what
experiences tell you that in this case, a secondary mechanism is needed? If the
failure mode of the outputs is to all turn off, no worries then. I'm more
concerned about a situation where power is on and it's running and some
condition causes the other (direction control line) output to change state,
even for a few mS. I'd like to foresee as many contingencies as possible.

I also need to know how outputs respond if power to the pic is lost. Are there
any pulses or transients that outputs have or do they just fall to 0 (in real
experience)?

Thanks,
Skip


More... (looser matching)
- Last day of these posts
- In 2008 , 2009 only
- Today
- New search...